Escherichia coli is the most common cause of urinary tract infections in this country and the long term objective of this project is to understand the molecular basis of its virulence properties. Bacterial binding to a digalactoside receptor (galabiose) found in the globoseries of glycolipids in uroepithelial cells is a crucial step in pyelonephritis and this process is frequently mediated by filamentous heteropolymeric structures known as P pili. The P pilus system of Escherichia coli is a powerful model for detailed analyses of chaperone-assisted assembly of specialized surface structures. This proposal blends the powerful genetics of the P pilus system with X ray crystollographic data and biochemical studies to investigate the chaperone-assisted protein-protein interactions that are necessary for the assembly of subunit proteins into pili having adhesins at their tips. The assembly of P pili requires the chaperone protein, PapD. The periplasmic PapD chaperone has been shown to form preassembly complexes with the adhesin, PapG, and with another minor pilus protein, PapE, and in its absence the pilus protein subunits are degraded or accumulate as precursor proteins with an uncleaved signal sequence. The known three dimensional structure of PapD will be used to design an extensive series of site-directed point mutations in the proposed active site. The point mutant alleles will be recombined into the pap operon in the chromosome of wild type E. coli and the effect of the mutation on piliation and receptor binding will be determined by electron microscopy and hemagglutination titer. The first mutation generated in this way illustrates the power of this approach; the mutant papD allele results in pili that are kinked and are easily dissociated from the cell surface. The fine molecular details of the PapD-mediated protein-protein interactions that are interrupted by the mutation will then be determined. The levels of each pilus protein subunit in the wild type and mutant pili will be quantitated by ELISA and their relative location in the pilus determined by immunogold EM. The stability of each subunit type in vivo and the ability of the point mutant and wild type PapD to mediate protein folding of the adhesin in vitro will be determined. IN vivo pulse chase experiments of cultures where pilus assembly has been synchronized will be used to follow PapD mediated protein-protein interactions required for pilus assembly. The formation of preassembly complexes will also be determined in HPLC and electrophoretic analyses in papD mutant and wild type backgrounds. The mechanism of action of PapD will be investigated in in vitro folding, oligomerization and pilus tip assembly models and by testing the ability of the chaperone to convert a nonbinding adhesin (partially unfolded) into a binding conformation. The information gained from these studies will make possible the design of compounds that interrupt pilus assembly thus blocking bacterial adherence to the urinary mucosa.